Optimal Pump Scheduling to Pressure Management for Large-Scale Water Distribution Systems

Abstract

Water loss is a global problem. Water loss in leakages in water distribution systems (WDSs) can be significantly reduced as pressures at outlets of pumping stations are optimized. Due to the introduction of binary variables for describing on/off operations of pumps, the optimal pressure management is formulated as a large-scale mixed integer nonlinear programming (MINLP) problem which is extremely difficult to be solved by MINLP algorithms. Many studies have employed meta-heuristic algorithms to solve such the MINLP problem, but only applied for small-scale networks and it took an expensive computation time. This paper develops an efficient formulation of MINLP for optimal pressure managing problem. It is due to the fact that the large-scale MINLP can be decomposed into small-scaled MINLPs, hence it can be solved efficiently by MINLP solvers based gradient methods in an reasonable computation time. To demonstrate the efficiency of our solution approach, the real-world water distribution systems in Thainguyen city in Vietnam will be considered for optimization of pressure management. The resulted pumping schedules lead to higher reduction of excessive pressures and water leakage amounts in comparison with those by the current pumping schedule.